Introduction

The use of fossil energy produces a large number of nitrogen oxides, which could cause acid rain, photochemical smog, and harm to human health (Hou et al. 2014; Jiang et al. 2016; Tang et al. 2016). With the implementation of ultra-low emission standard for atmospheric pollutants in China, the promotion and implement of flue gas denitrification (De-NOx) technologies will be forced and stressed (Xie et al. 2019). The selective catalytic reduction of NOx by NH3 (NH3-SCR) is a mature technology to solve NOx emissions. The catalyst is essential for NH3-SCR technology, and it determines the operating temperature and cost of the industrial SCR system. V2O5-WO3/TiO2 is extensively used as the SCR denitration catalyst due to its high catalytic activity and resistance to SO2 poisoning. However, the active window temperature of V2O5-WO3/TiO2 catalyst is higher (320–400 °C), and it is not suitable for low-temperature flue gas (below 150 °C) in iron and steel, cement, glass, and other industries (Yang et al. 2020). Therefore, it is important to research a highly efficient NH3-SCR catalyst at low temperature. Some catalysts show a high SCR activity in the range of 120–250 °C, such as MnOx/Al2O3, CuO/Al2O3, and Mn/TiO2 catalyst, but their catalytic activities easily are inhibited by SO2(Hou et al. 2014).

Developing a catalyst with high resistance to SO2 deactivation at low temperature is critical to the NH3-SCR process. Studies have shown that it is superior to TiO2 and SiO2 in terms of resistance to ammonium sulfate, activated carbon, or coke as a carrier (Han et al. 2018). Compared with the activated carbon, activated coke (AC) has the characteristics of low cost and high mechanical strength (Ye et al. 2020). AC has a large specific surface area, abundant surface functional groups, and strong adsorption capacity, which had been successfully applied to SO2 and Hg capture and NOx purification of flue gas in the steel industry (Liu and Liu 2013, Li et al. 2018, Li et al. 2019, Ye et al. 2020). Pure AC shows a low catalytic activity; thus, the method of introducing new active components (such as MnOx, CuO, or V2O5) is used to enhance its denitration activity (Marban et al. 2004; Boyano et al. 2008; Lazaro et al. 2009). Compared with these catalysts, AC-supported V2O5 (V2O5/AC) catalysts are more promising, because those still have good catalytic activity in the presence of low moisture and SO2 conditions (Sun et al. 2009; Wang et al. 2014). In order to further develop the V2O5/AC catalyst, the studies about NH3-SCR reaction mechanism and active site were carried, and the result has shown that the V2O5 species are the main sites for the adsorption and activation of NH3(Sun et al. 2009).

AC has characteristic of flammable at high temperature; thus, N2 atmosphere is used during loading V2O5 species (the decomposition of NH4VO3 into V2O5 occurs only when the temperature is over 500 °C). After loading V2O5 species, and calcined (pre-oxidized) under air atmosphere to activate the V2O5 and catalyst. According to the literature, pre-oxidized catalyst can enhance catalytic activity, compared with the non-pre-oxidized catalyst, and the pre-oxidized catalyst has better catalytic activity and stability (Zhu et al. 1999). However, there is no literature report on the relationship between pre-oxidation process and catalytic activity. In this work, catalysts were prepared by different pre-oxidation time and tested for their NH3-SCR performance at low temperature, and the SEM, BET, XPS, H2-TPR, and NH3-TPD characterizations were performed to explore the reasons for this phenomenon and construct the structure-activity relationship of the catalyst.

Experimental

Catalyst preparation

The V2O5/AC catalyst was prepared by the impregnation method. Generally, 5 g activated coke (40–80 mesh) and 0.3214 g NH4VO3(AR) were added into 100 mL deionized water under vigorously stirring for 1 h at room temperature. Subsequently, the mixture solution was stirred at 80 °C for 2 h followed by drying at 120 °C for 5 h and then calcined at 400 °C in a N2 atmosphere for 5 h to obtain a V2O5/AC catalyst. After that, the catalyst was pre-oxidized at 250 °C, the obtained V/AC catalysts was labeled as V/AC-X (X was pre-oxidization time, X=0, 3, 4, and 5). Among them, V/AC-0 means the preparation of the V2O5/AC catalyst without pre-oxidation.

Characterization

The Supplementary materials provide the particulars of characterization.

Catalyst activity tests

The efficiency of NOx catalytic removal was tested in a fixed-bed reactor (i.d. 7mm). The mass of catalyst was 0.7 g (40–80 mesh) in the test, the feeding gases were mixed by the gas mixer and then sent to the reactor, the total flow rate was 200 mL/min, and the inlet gas hourly space velocity (GHSV) was maintained at 12000 h−1. The inlet gas concentrations of NO and NH3 in the feeding gas both were 500 ppm, while O2 was 5%, and the N2 was balanced gas. Meanwhile, 5 vol % H2O (when used), 300 ppm SO2 (when used) was added in the test. The values of NOx removal efficiency can be calculated by following equation:

$$ {\mathrm{NO}}_x\ \mathrm{removal}\ \mathrm{efficiency}\ \left(\%\right)=\left({\left[{\mathrm{NO}}_x\right]}_{\mathrm{in}}-{\left[{\mathrm{NO}}_x\right]}_{\mathrm{out}}\right)/{\left[{\mathrm{NO}}_x\right]}_{\mathrm{in}}\times 100\% $$

Results and discussion

Catalyst performance

Fig. 1 exhibited the NOx removal efficiency of AC and V/AC-X (X= 0, 3, 4, and 5) catalysts. The result indicated that the NOx conversion decreased in the following order: V/AC-4 > V/AC-3> V/AC-5> V/AC-0>AC. The AC exhibited a low activity, the NOx conversion at 120 °C was only 20%, and it reduced with the increase of reaction temperature. This result was similar with research previously (Jing et al. 2014). After loading V2O5 species, the NOx removal efficiency of the V/AC catalyst increased significantly, suggesting that V2O5 is the main active species of NH3-SCR reaction. After pre-oxidizing, the catalytic activities had been further improved; for the V/AC-0, the NOx conversion was 76% at 240 °C, while V/AC-X were 85% (V/AC-3) and 94% (V/AC-4). However, the NOx conversion of V/AC-5 was 83% at 240 °C, which was lower than V/AC-4. In summary, the pre-oxidation process can contribute to the superior NH3-SCR performance of V/AC-X catalyst, and pre-oxidation time is the most important influencing factor; the best pre-oxidize time is 4 h.

Fig. 1
figure 1

NOx conversion of samples at different temperatures

Full size image

The long-term stability of V/AC-4 catalyst was investigated by testing the NH3-SCR performance of the catalyst at 240 °C for 60 h. The result was exhibited in Fig. S1a. The NOx removal efficiency of the catalyst was not decreased within 60 h, but increased a little (the NOx conversion has increased from 94.0 to 97.7%), suggesting that the stability of V/AC-4 catalyst was good. The effect of SO2 and water vapor on V/AC-4 catalyst was examined at 240 °C. As presented in Fig. S1b–c, when 300 ppm SO2 and 5 vol% H2O were added, respectively, the NOx removal efficiency of V/AC catalyst increases and keeps the increase trend slightly. After removing SO2 and H2O, the catalytic activity of V/AC-4 is still increased with time, indicating that the tolerance of SO2 and H2O over V/AC-4 catalyst was good. After removing SO2, the catalytic activity is still increased with time. The reason for the increased activity could be caused by the sulfate formed during test process, which enhanced the surface acidity of catalyst (Zhu et al. 2001). The more surface acidity of the catalyst could be beneficial to the adsorption and activation of ammonia, thereby increasing the NH3-SCR activity. In general, water vapor and reaction gas produce competitive adsorption on the catalyst surface (Hou et al. 2014; Lu et al. 2018), thereby inhibiting the adsorption of reactive species and reducing the NH3-SCR performance. Due to the hydrophobicity of the activated coke (Zhang et al. 2020), water vapor cannot adsorb on the catalyst surface, explaining why the V/AC-4 catalysts showed good H2O resistance.

Catalyst characterization

In order to explain the influence of pre-oxidation process on the crystal structures of the catalyst, the AC and V/AC-X catalysts were characterized by XRD, respectively. The results are presented in Fig. 2. Two broad diffraction peaks located at 20–30° and 40–50° are characteristic diffractions of activated coke graphite crystals (Han et al. 2018; Yang et al. 2020). The characteristic peaks of SiO2 (an additive substance used in the industrial production of activated coke) located at 20.8°, 26.6°, 36.6°, 44.0°, 50.1°, 60.0°, and 68.2° were observed (Ge et al. 2019). No peaks that belong to V2O5 species were detected in the patterns of V/AC catalyst, suggesting that vanadium species has extremely high dispersibility on the surface of AC carrier.

Fig. 2
figure 2

XRD patterns of catalysts

Full size image

The physical structures of AC and V/AC-X catalysts are presented in Table 1. AC exhibited a specific surface area of 214 m2 g−1 and a total pore volume of 0.025 cm3 g−1 with an average pore diameter of 4.44 nm. Compared with AC catalyst, the BET surface area, pore volume, and average pore size of the V/AC-0 catalyst decreased. With the increase of the pre-oxidation time, the specific surface area, pore volume, and average pore size of the V/AC-3 and V/AC-4 catalysts both increased, while the V/AC-5 catalyst is greatly reduced, which is consistent with activity test results, which due to the structure of V/AC-X catalyst changed during the pre-oxidation process, thereby affecting the catalytic activity. Compared with the V/AC-0 catalyst, the surface area of V/AC-5 catalyst is smaller, while the catalytic activity is better, indicating that the specific surface area is not the main factor for the enhancement of the catalytic activity. The specific surface area of the V/AC-4 catalyst is similar to the V/AC-3 and V/AC-0 catalyst, while the pore volume and average pore size of V/AC-4 catalyst was the largest, and it had the best NH3-SCR performance, thus indicating that the new pore structure is formed to facilitate the NH3-SCR reaction. In order to further explore the impact of the pre-oxidation process on the catalyst structure, the sample morphology and particle size were analyzed by SEM, and the results are presented in Fig. S2. Compared with AC catalyst, the surface of V/AC-0 catalyst is uneven (Fig. S2a–b), and it has shown that the structure of the catalyst was changed after loading vanadium species. With the increase of the pre-oxidation time, there was agglomerative phenomenon accompanying with V/AC-3 and V/AC-4 catalyst (Fig. S2c–e), while the V/AC-5 catalyst presents a number of large block crystals, obviously, which is consistent with the previous BET results. This phenomenon can explain the reason for that the NOx removal efficiency of V/AC-5 catalyst is lower than that the V/AC-4 catalyst. Both results of SEM and BET tests have shown that the new pore structure formed in the catalyst surface during the pre-oxidation process, thus affecting their SCR performance at low temperature.

Table 1 Specific surface area, pore volume, and pore diameter of the catalysts
Full size table

The surface elemental compositions of AC, V/AC-X, and V/AC-4-used (after 60-h stability test) catalysts were tested by XPS. The V2p1/2 XPS results are exhibited in Fig. 3a. The peaks at 515.7–517.0 eV and 517.0–518.0 eV were attributed to V4+ and V5+, respectively (Cha et al. 2013). The content of V4+ on the catalyst was calculated by the peak area V4+/(V4++V5+). As presented in Table 2, the content of V4+ on V/AC-0, V/AC-3, V/AC-4, and V/AC-5 catalysts was 60.6%, 43.3%, 30.6%, and 29.0%, respectively. With the increase of the pre-oxidation time, the V4+ content decreases, which indicated that the V4+ was oxidized to V5+ by the oxygen in air atmosphere (Zhu et al. 1999). After stability testing, the V4+ content of V/AC-4 sample increased from 30.6% (V/AC-4-fresh) to 32.4% (V/AC-4-used), which is due to the reduction of V5+ into V4+ by NH3 during NH3-SCR reaction.

Fig. 3
figure 3

The XPS spectra of catalysts: a V 2p1/2 and b O 1s

Full size image
Table 2 Element composition percentage
Full size table

The O1s XPS spectra are exhibited in Fig. 3b. The peaks at 529–530 eV, 531–533 eV, and 533–535 eV were attributed to lattice oxygen (marked as Oβ), surface chemical oxygen (marked as Oα), and the oxygen species in hydroxide groups (marked as Oγ), respectively (Jiang et al. 2019). The proportions of Oα/(Oα+ Oβ+ Oγ) in V/AC-0 sample was 20.88%. After pre-oxidizing, the content of Oα is significantly increased. The content of Oα of V/AC-3, V/AC-4, and V/AC-5 catalysts were 48.22%, 49.05%, and 48.32%. According to previous reports, chemically adsorbed oxygen can boost that the NO was oxidized to NO2, which can promote “fast SCR” reaction (Ma et al. 2011). After pre-oxidizing, the V4+ content decreased, and the Oα content increased, indicating that the V4+ was oxidized by oxygen molecule during pre-oxidation process, which contributes to the formation of V5+ ions and surface-active oxygen species, thereby improving the catalytic activity. Compared with the V/AC-4 catalyst, the Oα content and V4+ content of V/AC-4-used sample were both increased in the test, indicating that the Oα was formed during the V4+ and was oxidized to V5+, while V4+ was generated again in NH3-SCR reaction atmosphere, which can enhance the cycle of V4+/V5+, which explains why the catalytic activity increased gradually in the stability test. With the increase of the pre-oxidation time, the Oγ content increases, and this may be due to the formation of hydroxyl functional groups during the pre-oxidation process. The catalytic activity of the V/AC-3 catalyst is better than the V/AC-5 catalyst, which is different from the activity results, thereby the content of Oγ is not the influencing factor of the catalytic activity.

The redox property of catalyst is one of the important influencing factors in SCR reaction. Fig. 4 exhibited the H2-TPR curves of AC and V/AC-X catalysts. The peak located at 624 °C for AC catalyst was attributed to the vaporization peak of carbon carrier on activated coke (Zhang et al. 2020). The peaks located at 624 °C and 513 °C in the V/AC-0 catalyst were caused by the vaporization peak of carbon carrier and reduction peak of V5+, respectively (Han et al. 2018). Compared with the AC catalyst, the reduction peak (after pre-oxidation) of V/AC-X catalyst shifted to a lower temperature, indicating that the interaction between activated coke carrier and vanadium species could improve the redox property of catalysts (Lu et al. 2018). With the increase of pre-oxidation time, the intensity of the reduction peak of V5+ and the vaporization peak of the activated coke carrier of the V/AC-3 and V/AC-4 catalyst are obviously enhanced, indicating that the content of V5+ and the redox sites of the catalyst are both increases during the pre-oxidation progresses. However, the reduction peak area of the V/AC-5 catalyst is significantly reduced. Combined with the result of BET, it could be shown that the excessive pre-oxidation time causes the sintering of active species on V/AC-5 catalyst, thereby decreasing its catalytic activity.

Fig. 4
figure 4

H2-TPR spectra of catalysts

Full size image

NH3-TPD tests are performed to study the ammonia desorption over AC, V/AC-X, and V/AC-4-SO2 (after sulfur resistance test) samples. As shown in Fig. 5a, the AC curve showed only one peak at 147 °C, while the V/AC-X samples present two broad peaks; the desorption peaks located at 147 °C and 170 °C were the physical adsorption of ammonia and the adsorption of NH4+ at the Brønsted acid site (Yin et al. 2018). After loading vanadium species, the desorption peak area increased significantly, indicating that vanadium species can provide more acidic sites on the catalyst surface, which can promote the adsorption and activation of ammonia. Compared with V/AC-0 catalyst, the desorption peak areas of the V/AC-3 and V/AC-4 further increase, suggesting that the pre-oxidation process promotes the adsorption of ammonia and can promote the formation of V5+ on the catalyst surface (Xu et al. 2018), which is consistent with the H2-TPR results. However, the desorption peak area of the V/AC-5 catalyst is reduced, which is because catalytic activity species was agglomerated, and the adsorption of NH3 is inhibited. As shown in Fig. 5b, NH3-TPD patterns were performed among V/AC-4-SO2 (after sulfur resistance test) samples, the desorption peak area of V/AC-4-SO2 sample is larger, and this is because the sulfate formed during test process could improve the adsorption and activation of ammonia, thereby increasing the NH3-SCR activity.

Fig. 5
figure 5

NH3-TPD spectra of catalysts

Full size image

Conclusion

The activity test results showed that the NOx catalytic removal efficiency of V2O5/AC catalyst is better than the AC catalyst, suggesting that the V2O5 is the active substance. The activity sequence of V/AC-X catalysts prepared by different pre-oxidation time at 120–240 °C is V/AC-4> V/AC-3> V/AC-5> V/AC-0. The results showed that the pre-oxidation process is beneficial to catalytic performance of V2O5/AC-X catalysts. Series of characterizations were used to explain the reasons for the phenomenon. The BET and SEM characterization results demonstrated that the specific surface area, pore volume, and average pore size of the catalyst increase with the increase of the pre-oxidation time, and the new pore structure is formed to facilitate the NH3-SCR reaction. During the pre-oxidation progress, the V4+ was oxidized by oxygen molecule, which contributes to the formation of V5+ ions and surface-active oxygen species. The surface-active oxygen species is conducive to promoting the “fast SCR.” In summary, the pre-oxidized V2O5/AC has the best catalytic activity when the pre-oxidation time is 4 h.